The AURORA project: ScientificDevelopmentsand Results

The Advanced Ultraviolet Radiation and Ozone Retrieval for Applications (AURORA) is a three-year project supported by the European Union during the 2016-2019 period. The main goal of AURORA is to develop and test the coupling of a data fusion algorithm with state-of-the-art data assimilation systems in an end-to-end infrastructure that could be used as prototype for the exploitation of future Copernicus Sentinel-4 (S4) and Sentinel-5 (S5) atmospheric data. A simplified Observing System Simulated Experiment approach has been adopted in the project. Using MERRA-2 reanalysis as virtual truth, synthetic measurements have been generated to simulate the future S4 and S5 data in three spectral ranges (i.e. Ultraviolet, Visible and Thermal Infrared). Level 2 ozone products have then been derived for both platforms and the three spectral ranges. A data fusion technique, referred to as the Complete Data Fusion (CDF), has been applied as an a-posteriori method to combine the Level-2 products into a comprehensive and concise description of that atmospheric state. Both the Level-2 and fused ozone products have been assimilated in two state-of-the-art data assimilation systems: the ECMWF IFS and KNMI TM5. An incremental assimilation experiment design has been defined to assess the potential value of (1) assimilating fused data instead of the Level-2 retrievals, (2) using fused data from one geostationary platform in addition to polar orbiting-based fused data, and (3) exploiting cross-platform fused data instead of jointly assimilating fused products from multiple platforms. A collaboration with the American TEMPO and Korean GEMS geostationary satellite communities has been established in the AURORA project with the aim of evaluating the value of increasing the geostationary coverage from one (S4 only) to three. Off-line tropospheric ozone and surface UV products are also generated from the data assimilation ozone analyses and forecasts. A validation function has been developed to assess the quality of all the AURORA outputs against both independent observations and the MERRA-2 reanalysis used to derive the simulated measurements. This contribution will present the scientific developments undertaken within the AURORA project, and summarize its results.